Anomaly detection in time series based on interval sets

Anomaly detection in time series is a popular topic focusing on a variety of applications, which achieves a wealth of results. However, there are many cases of missing anomalies and false alarms for most existing works. Inspired by the concept of interval sets, this paper proposes an anomaly detection algorithm called probability interval and tries to detect the anomaly data in time series from a new perspective. In the proposed algorithm, a time series is divided into several subsequences. Each subsequence is regarded as an interval set depending on its value space and boundary of the subsequence. The similarity measurements between interval sets adopt interval operations and point probability distributions of the interval bounds. In addition, based on similarity results, the anomaly score is defined. The experimental results on artificial and real datasets indicate that the proposed algorithm has better discriminative performance than the piecewise aggregate approximation method and greatly reduces the false alarm rate.